Foundations of Data Science
In the Summer Semester 2019, I was responsible for tutoring and organizing the lecture Foundations of Data Science at Skoltech
In this course, we introduce the forefront of modern research in data science and familiarize Ph.D. students with state of the art in those areas. In particular, we introduce cornerstone subjects that are not commonly discussed in undergraduate or graduate Machine Learning classes. This course explores extensively three different areas in Machine Learning, namely Causality, Topological Data Analysis, Graph neural networks, Optimal Transport, and Deep Learning Theory.
Over multiple weeks, we will investigate how these methods and algorithms can be used for analyzing scientific data, social networks, or time-series data, creating recommender systems, mining sequences, carrying out text/web analysis, topic modeling, and pattern mining.
We explore how these concepts are applied for dimensionality reduction and manifold learning, combinatorial optimization, relational and structured learning, classification and regression methods, semi-supervised learning, unsupervised learning including anomaly detection and clustering, kernel methods, compressed sensing and sparse modeling, Bayesian methods, deep learning, hyper-parameter and model selection, Markov decision processes, reinforcement learning, dynamical systems and Hidden Markov Processes.
The course aims to bring all students on the same page regarding the nature and orientation of state-of-the-artwork in their field so that they acquire both depth and breadth of knowledge.